Breast cancer cells often exhibit functional and phenotypic heterogeneity. Cellular heterogeneity may be the source of drug resistance and dormancy promoting cancer relapse. One source of tumor cell heterogeneity may be the existence of cancer stem cells. Based on this hypothesis, the tumorigenic CSCs, herein referred to as tumor initiating cells (TICs), differentiate into non-tumorigenic cancer cells creating a hierarchical organization and tumor cell heterogeneity. Breast TICs have been identified by the expression of unique extracellular surface markers such as CD44hi/CD24lo. It is now believed that even TICs may exhibit cellular heterogeneity through clonal evolution. In order to study heterogeneity among breast TICs, we utilized Raman Spectroscopy (RS). RS enables the characterization of intracellular molecular composition of live TICs. By performing RS, we demonstrated that live TICs exhibited both inter- and intra-tumoral heterogeneity based on the expression of a unique intracellular molecular component. Based on these data, we hypothesize that TICs are heterogeneous and Raman Tweezer Spectroscopy (RTS) may be a method by which to sort and functionally evaluate the tumorigenic potential of TICs groups that exhibit unique intracellular molecular profiles (IMPs). Our long-term goal is to identify and eradicate the most resistant, dormant, and aggressive breast TICs in order to prevent cancer relapse. Our objective is to characterize heterogeneity among TICs by IMP using RTS. Our Rationale is that so far, TICs have been isolated and characterized by the expression of unique surface markers. However, a methodology for isolation and characterization of TICs based on combined extracellular and IMP has not been evaluated. RS is a vibrational spectroscopy. Photons of the laser light are absorbed by live cells and then reemitted, providing spectral profiles (wavelengths) that represent the intracellular biochemical signatures of living cells. RTS will be combined with advanced mathematical tools to quantify characteristic IMPs of TICs non-invasively, with high specificity and rapidly. RTS permits sorting cells according to their IMPs. The trapped cells will then be functionally evaluated for their tumorigenic potential by limiting dilution transplantation assays. These studies will provide the basis for future molecular characterization of the most aggressive and indolent breast TICs and design of therapeutic strategies for their eradiation.